Pandas Profiling Report

Dataset statistics

Number of variables	7
Number of observations	4000
Missing cells	2223
Missing cells (%)	7.9%
Duplicate rows	0
Duplicate rows (%)	0.0%
Total size in memory	218.9 KiB
Average record size in memory	56.0 B

Variable types

NUM	4
CAT	2
BOOL	1

Warnings

`Height` has 1894 (47.4%) missing values	Missing
`Weight` has 326 (8.2%) missing values	Missing
`PATIENT_ID` has unique values	Unique

Reproduction

Analysis started	2020-09-17 13:46:38.843291
Analysis finished	2020-09-17 13:46:48.004032
Duration	9.16 seconds
Software version	pandas-profiling v2.9.0
Download configuration	config.yaml

PATIENT_ID
Real number (ℝ_≥0)

UNIQUE

Distinct	4000
Distinct (%)	100.0%
Missing	0
Missing (%)	0.0%
Infinite	0
Infinite (%)	0.0%

Mean	137605.122
Minimum	132539
Maximum	142673
Zeros	0
Zeros (%)	0.0%
Memory size	31.2 KiB

Quantile statistics

Minimum	132539
5-th percentile	133038.95
Q1	135075.75
median	137592.5
Q3	140100.25
95-th percentile	142176.2
Maximum	142673
Range	10134
Interquartile range (IQR)	5024.5

Descriptive statistics

Standard deviation	2923.608886
Coefficient of variation (CV)	0.02124636673
Kurtosis	-1.191489871
Mean	137605.122
Median Absolute Deviation (MAD)	2513
Skewness	0.00584744361
Sum	550420488
Variance	8547488.92
Monotocity	Strictly increasing

Histogram with fixed size bins (bins=50)

Value	Count	Frequency (%)
135162	1	< 0.1%
139889	1	< 0.1%
142205	1	< 0.1%
141960	1	< 0.1%
133764	1	< 0.1%
135811	1	< 0.1%
139905	1	< 0.1%
142106	1	< 0.1%
135803	1	< 0.1%
137850	1	< 0.1%
Other values (3990)	3990	99.8%

Minimum 5 values
Maximum 5 values

Value	Count	Frequency (%)
132539	1	< 0.1%
132540	1	< 0.1%
132541	1	< 0.1%
132543	1	< 0.1%
132545	1	< 0.1%

Value	Count	Frequency (%)
142673	1	< 0.1%
142671	1	< 0.1%
142670	1	< 0.1%
142667	1	< 0.1%
142665	1	< 0.1%

ihd
Boolean

Distinct	2
Distinct (%)	< 0.1%
Missing	0
Missing (%)	0.0%
Memory size	31.2 KiB

0	3446
1	554

Common Values
Chart

Value	Count	Frequency (%)
0	3446	86.2%
1	554	13.9%

Age
Real number (ℝ_≥0)

Distinct	76
Distinct (%)	1.9%
Missing	0
Missing (%)	0.0%
Infinite	0
Infinite (%)	0.0%

Mean	64.2475
Minimum	15
Maximum	90
Zeros	0
Zeros (%)	0.0%
Memory size	31.2 KiB

Quantile statistics

Minimum	15
5-th percentile	29
Q1	52.75
median	67
Q3	78
95-th percentile	89
Maximum	90
Range	75
Interquartile range (IQR)	25.25

Descriptive statistics

Standard deviation	17.56094646
Coefficient of variation (CV)	0.2733327594
Kurtosis	-0.3325881438
Mean	64.2475
Median Absolute Deviation (MAD)	12
Skewness	-0.6062680114
Sum	256990
Variance	308.3868405
Monotocity	Not monotonic

Histogram with fixed size bins (bins=50)

Value	Count	Frequency (%)
90	164	4.1%
77	126	3.1%
78	104	2.6%
83	102	2.5%
79	100	2.5%
74	93	2.3%
72	90	2.2%
80	89	2.2%
73	88	2.2%
81	86	2.1%
Other values (66)	2958	74.0%

Minimum 5 values
Maximum 5 values

Value	Count	Frequency (%)
15	1	< 0.1%
16	1	< 0.1%
17	4	0.1%
18	12	0.3%
19	18	0.4%

Value	Count	Frequency (%)
90	164	4.1%
89	41	1.0%
88	51	1.3%
87	49	1.2%
86	65	1.6%

Gender
Categorical

Distinct	2
Distinct (%)	0.1%
Missing	3
Missing (%)	0.1%
Memory size	31.2 KiB

male	2246
female	1751

Frequencies
Length

Value	Count	Frequency (%)
male	2246	56.1%
female	1751	43.8%
(Missing)	3	0.1%

Frequencies of value counts

Unique

Unique	0 ?
Unique (%)	0.0%

Histogram of lengths of the category

Length

Max length	6
Median length	4
Mean length	4.87475
Min length	3

Height
Real number (ℝ_≥0)

MISSING

Distinct	70
Distinct (%)	3.3%
Missing	1894
Missing (%)	47.4%
Infinite	0
Infinite (%)	0.0%

Mean	169.787227
Minimum	1.8
Maximum	431.8
Zeros	0
Zeros (%)	0.0%
Memory size	31.2 KiB

Quantile statistics

Minimum	1.8
5-th percentile	152.4
Q1	162.6
median	170.2
Q3	177.8
95-th percentile	185.4
Maximum	431.8
Range	430
Interquartile range (IQR)	15.2

Descriptive statistics

Standard deviation	20.17460395
Coefficient of variation (CV)	0.1188228603
Kurtosis	77.10913097
Mean	169.787227
Median Absolute Deviation (MAD)	7.6
Skewness	3.303175323
Sum	357571.9
Variance	407.0146444
Monotocity	Not monotonic

Histogram with fixed size bins (bins=50)

Value	Count	Frequency (%)
177.8	213	5.3%
182.9	188	4.7%
170.2	173	4.3%
172.7	169	4.2%
167.6	162	4.0%
162.6	153	3.8%
157.5	142	3.5%
175.3	137	3.4%
165.1	124	3.1%
180.3	115	2.9%
Other values (60)	530	13.2%
(Missing)	1894	47.3%

Minimum 5 values
Maximum 5 values

Value	Count	Frequency (%)
1.8	1	< 0.1%
13	3	0.1%
13.7	1	< 0.1%
14	2	0.1%
15.2	1	< 0.1%

Value	Count	Frequency (%)
431.8	1	< 0.1%
426.7	1	< 0.1%
419.1	1	< 0.1%
406.4	1	< 0.1%
398.8	1	< 0.1%

ICUType
Categorical

Distinct	4
Distinct (%)	0.1%
Missing	0
Missing (%)	0.0%
Memory size	31.2 KiB

Medical ICU	1481
Surgical ICU	1068
Cardiac Surgery Recovery Unit	874
Coronary Care Unit	577

Frequencies
Length

Value	Count	Frequency (%)
Medical ICU	1481	37.0%
Surgical ICU	1068	26.7%
Cardiac Surgery Recovery Unit	874	21.9%
Coronary Care Unit	577	14.4%

Frequencies of value counts

Unique

Unique	0 ?
Unique (%)	0.0%

Histogram of lengths of the category

Length

Max length	29
Median length	12
Mean length	16.20975
Min length	11

Weight
Real number (ℝ_≥0)

MISSING

Distinct	836
Distinct (%)	22.8%
Missing	326
Missing (%)	8.2%
Infinite	0
Infinite (%)	0.0%

Mean	81.47826892
Minimum	21.7
Maximum	300
Zeros	0
Zeros (%)	0.0%
Memory size	31.2 KiB

Quantile statistics

Minimum	21.7
5-th percentile	50.5
Q1	66
median	78.6
Q3	92
95-th percentile	122
Maximum	300
Range	278.3
Interquartile range (IQR)	26

Descriptive statistics

Standard deviation	23.62843246
Coefficient of variation (CV)	0.2899967411
Kurtosis	7.409920777
Mean	81.47826892
Median Absolute Deviation (MAD)	13
Skewness	1.704291978
Sum	299351.16
Variance	558.3028205
Monotocity	Not monotonic

Histogram with fixed size bins (bins=50)

Value	Count	Frequency (%)
70	98	2.5%
80	79	2.0%
90	68	1.7%
60	61	1.5%
65	56	1.4%
75	54	1.4%
100	53	1.3%
85	43	1.1%
77	32	0.8%
82	31	0.8%
Other values (826)	3099	77.5%
(Missing)	326	8.2%

Minimum 5 values
Maximum 5 values

Value	Count	Frequency (%)
21.7	1	< 0.1%
31.7	1	< 0.1%
32	1	< 0.1%
34.6	1	< 0.1%
35	1	< 0.1%

Value	Count	Frequency (%)
300	1	< 0.1%
280	1	< 0.1%
253	1	< 0.1%
230	1	< 0.1%
220	3	0.1%

Pearson's r

The Pearson's correlation coefficient (r) is a measure of linear correlation between two variables. It's value lies between -1 and +1, -1 indicating total negative linear correlation, 0 indicating no linear correlation and 1 indicating total positive linear correlation. Furthermore, r is invariant under separate changes in location and scale of the two variables, implying that for a linear function the angle to the x-axis does not affect r.

To calculate r for two variables X and Y, one divides the covariance of X and Y by the product of their standard deviations.

Spearman's ρ

The Spearman's rank correlation coefficient (ρ) is a measure of monotonic correlation between two variables, and is therefore better in catching nonlinear monotonic correlations than Pearson's r. It's value lies between -1 and +1, -1 indicating total negative monotonic correlation, 0 indicating no monotonic correlation and 1 indicating total positive monotonic correlation.

To calculate ρ for two variables X and Y, one divides the covariance of the rank variables of X and Y by the product of their standard deviations.

Kendall's τ

Similarly to Spearman's rank correlation coefficient, the Kendall rank correlation coefficient (τ) measures ordinal association between two variables. It's value lies between -1 and +1, -1 indicating total negative correlation, 0 indicating no correlation and 1 indicating total positive correlation.

To calculate τ for two variables X and Y, one determines the number of concordant and discordant pairs of observations. τ is given by the number of concordant pairs minus the discordant pairs divided by the total number of pairs.

Phik (φk)

Phik (φk) is a new and practical correlation coefficient that works consistently between categorical, ordinal and interval variables, captures non-linear dependency and reverts to the Pearson correlation coefficient in case of a bivariate normal input distribution. There is extensive documentation available here.

Cramér's V (φc)

Cramér's V is an association measure for nominal random variables. The coefficient ranges from 0 to 1, with 0 indicating independence and 1 indicating perfect association. The empirical estimators used for Cramér's V have been proved to be biased, even for large samples. We use a bias-corrected measure that has been proposed by Bergsma in 2013 that can be found here.

First rows

	PATIENT_ID	ihd	Age	Gender	Height	ICUType	Weight
0	132539	0	54	female	NaN	Surgical ICU	NaN
1	132540	0	76	male	175.3	Cardiac Surgery Recovery Unit	76.0
2	132541	0	44	female	NaN	Medical ICU	56.7
3	132543	0	68	male	180.3	Medical ICU	84.6
4	132545	0	88	female	NaN	Medical ICU	NaN
5	132547	0	64	male	180.3	Coronary Care Unit	114.0
6	132548	0	68	female	162.6	Medical ICU	87.0
7	132551	1	78	female	162.6	Medical ICU	48.4
8	132554	0	64	female	NaN	Medical ICU	60.7
9	132555	0	74	male	175.3	Cardiac Surgery Recovery Unit	66.1

Last rows

	PATIENT_ID	ihd	Age	Gender	Height	ICUType	Weight
3990	142655	0	43	male	NaN	Medical ICU	92.9
3991	142659	0	88	male	NaN	Coronary Care Unit	90.7
3992	142661	0	89	male	177.8	Surgical ICU	64.0
3993	142662	0	86	male	162.6	Medical ICU	53.0
3994	142664	0	51	female	NaN	Surgical ICU	75.0
3995	142665	0	70	female	NaN	Surgical ICU	87.0
3996	142667	0	25	male	NaN	Medical ICU	166.4
3997	142670	0	44	male	NaN	Medical ICU	109.0
3998	142671	1	37	male	NaN	Medical ICU	87.4
3999	142673	0	78	female	157.5	Surgical ICU	70.7

Overview

Variables

Interactions

Correlations

Pearson's r

Spearman's ρ

Kendall's τ

Phik (φk)

Cramér's V (φc)

Missing values

Sample

First rows

Last rows